Chapter6-7

Chapter6 Scales, axes and legends

6.1 Introduction

6.2 Modifying scales

library(ggplot2)
ggplot(mpg, aes(displ, hwy)) +
  geom_point(aes(colour = class))

##Default
ggplot(mpg, aes(displ, hwy)) +
  geom_point(aes(colour = class)) +
  scale_x_continuous() +
  scale_y_continuous() +
  scale_colour_discrete()

##override some default  
ggplot(mpg, aes(displ, hwy)) +
  geom_point(aes(colour = class)) +
  scale_x_continuous("A really awesome x axis label") +
  scale_y_continuous("An amazingly great y axis label")

ggplot(mpg, aes(displ, hwy)) +
  geom_point() +
  scale_x_continuous("Label 1") +
  scale_x_continuous("Label 2")

## Scale for 'x' is already present. Adding another scale for 'x', which
## will replace the existing scale.

ggplot(mpg, aes(displ, hwy)) +
  geom_point() +
  scale_x_continuous("Label 2")

ggplot(mpg, aes(displ, hwy)) +
  geom_point(aes(colour = class)) +
  scale_x_sqrt() +
  scale_colour_brewer()

scale_The name of the aesthetic (e.g., colour, shape or x)

or

scale_The name of the scale (e.g., continuous, discrete, brewer).

6.3 Guides: legends and axes

Guides allow you to read observations from the plot and map them back to their original values.

6.3.1 Scale title

df <- data.frame(x = 1:2, y = 1, z = "a")
p <- ggplot(df, aes(x, y)) + geom_point()
p + scale_x_continuous("X axis")

p + scale_x_continuous(quote(a + mathematical ^ expression))

p <- ggplot(df, aes(x, y)) + geom_point(aes(colour = z))
p +
  xlab("X axis") +
  ylab("Y axis")

p + labs(x = "X axis", y = "Y axis", colour =
           "Colour\nlegend")

## no label
p <- ggplot(df, aes(x, y)) + 
  geom_point() +
  theme(plot.background = element_rect(colour = "grey50"))
p + labs(x = "", y = "")

p + labs(x = NULL, y = NULL)

6.3.2 Breaks and labels

The breaks argument controls which values appear as tick marks on axes and keys on legends.

df <- data.frame(x = c(1, 3, 5) * 1000, y = 1)
axs <- ggplot(df, aes(x, y)) +
  geom_point() +
  labs(x = NULL, y = NULL)
axs

axs + scale_x_continuous(breaks = c(2000, 4000))

axs + scale_x_continuous(breaks = c(2000, 4000), labels = c("2k", "4k"))  

leg <- ggplot(df, aes(y, x, fill = x)) +
  geom_tile() +
  labs(x = NULL, y = NULL)
leg

leg + scale_fill_continuous(breaks = c(2000, 4000))

leg + scale_fill_continuous(breaks = c(2000, 4000), labels = c("2k", "4k"))

If you want to relabel the breaks in a categorical scale, you can use a named labels vector:

df2 <- data.frame(x = 1:3, y = c("a", "b", "c"))
ggplot(df2, aes(x, y)) +
  geom_point()

ggplot(df2, aes(x, y)) +
  geom_point() +
  scale_y_discrete(labels = c(a = "apple", b = "banana", c = "carrot"))

axs + scale_x_continuous(breaks = NULL)

axs + scale_x_continuous(labels = NULL)

leg + scale_fill_continuous(breaks = NULL)

leg + scale_fill_continuous(labels = NULL)

The scales package provides a number of useful labelling functions:

axs + scale_y_continuous(labels = scales::percent_format())

axs + scale_y_continuous(labels = scales::dollar_format("$"))

leg + scale_fill_continuous(labels = scales::unit_format("k", 1e-3))

df <- data.frame(x = c(2, 3, 5, 10, 200, 3000), y = 1)
ggplot(df, aes(x, y)) +
  geom_point() +
  scale_x_log10()

mb <- as.numeric(1:10 %o% 10 ^ (0:4))
ggplot(df, aes(x, y)) +
  geom_point() +
  scale_x_log10(minor_breaks = log10(mb))

6.4 Legends

6.4.1 Layers and legends

df <- data.frame(x=0,y = c(1, 2, 3), z=c('a','b','c'))
ggplot(df, aes(y, y)) +
  geom_point(size = 4, colour = "grey20") +
  geom_point(aes(colour = z), size = 2)

ggplot(df, aes(y, y)) +
  geom_point(size = 4, colour = "grey20", show.legend = TRUE) +
  geom_point(aes(colour = z), size = 2)

norm <- data.frame(x = rnorm(1000), y = rnorm(1000))
norm$z <- cut(norm$x, 3, labels = c("a", "b", "c"))
ggplot(norm, aes(x, y)) +
  geom_point(aes(colour = z), alpha = 0.1)

ggplot(norm, aes(x, y)) +
  geom_point(aes(colour = z), alpha = 0.1) +
  guides(colour = guide_legend(override.aes = list(alpha = 1)))

ggplot(df, aes(x, y)) + geom_point(aes(colour = z))

ggplot(df, aes(x, y)) + geom_point(aes(shape = z))

ggplot(df, aes(x, y)) + geom_point(aes(shape = z, colour = z))

6.4.2 Legend layout

df <- data.frame(x = 1:3, y = 1:3, z = c("a", "b", "c"))
base <- ggplot(df, aes(x, y)) +
  geom_point(aes(colour = z), size = 3) +
  xlab(NULL) +
  ylab(NULL)
base + theme(legend.position = "right") # the default

base + theme(legend.position = "bottom")

base + theme(legend.position = "none")

legend.direction: layout of items in legends (“horizontal” or “vertical”) legend.box: arrangement of multiple legends (“horizontal” or “vertical”) legend.box.just: justification of each legend within the overall bounding box, when there are multiple legends (“top”, “bottom”, “left”, or “right”)

base <- ggplot(df, aes(x, y)) +
  geom_point(aes(colour = z), size = 3)
base + theme(legend.position = c(0, 1), legend.justification = c(0, 1))

base + theme(legend.position = c(0.5, 0.5), legend.justification = c(0.5, 0.5))

base + theme(legend.position = c(1, 0), legend.justification = c(1, 0))

6.4.3 Guide functions

df <- data.frame(x = 1, y = 1:3, z = 1:3)
base <- ggplot(df, aes(x, y)) + geom_raster(aes(fill = z))
base

base + scale_fill_continuous(guide = guide_legend())

base + guides(fill = guide_legend())

6.4.3.1 guide_legend()

The legend guide displays individual keys in a table.

df <- data.frame(x = 1, y = 1:4, z = letters[1:4])
# Base plot
p <- ggplot(df, aes(x, y)) + geom_raster(aes(fill = z))
p

p + guides(fill = guide_legend(ncol = 2))

p + guides(fill = guide_legend(ncol = 2, byrow = TRUE))

p <- ggplot(df, aes(1, y)) + geom_bar(stat = "identity", aes(fill = z))
p

p + guides(fill = guide_legend(reverse = TRUE))

6.4.3.2 guide_colourbar

df <- data.frame(x = 1, y = 1:4, z = 4:1)
p <- ggplot(df, aes(x, y)) + geom_tile(aes(fill = z))
p

p + guides(fill = guide_colorbar(reverse = TRUE))

p + guides(fill = guide_colorbar(barheight = unit(4, "cm")))

6.5 Limits

df <- data.frame(x = 1:3, y = 1:3)
base <- ggplot(df, aes(x, y)) + geom_point()
base

base + scale_x_continuous(limits = c(1.5, 2.5))

## Warning: Removed 2 rows containing missing values (geom_point).

#> Warning: Removed 2 rows containing missing values (geom_point).
base + scale_x_continuous(limits = c(0, 4))

base + xlim(0, 4)

base + xlim(4, 0)

base + lims(x = c(0, 4))

To eliminate this space, set expand = c(0, 0). This is useful in conjunction with geom_raster():

ggplot(faithfuld, aes(waiting, eruptions)) + 
  geom_raster(aes(fill = density)) +
  theme(legend.position = "none")

ggplot(faithfuld, aes(waiting, eruptions)) +
  geom_raster(aes(fill = density)) +
  scale_x_continuous(expand = c(0,0)) +
  scale_y_continuous(expand = c(0,0)) +
  theme(legend.position = "none")

The default is scales::censor() which replaces any value outside the limits with NA. Another option is scales::squish() which squishes all values into the range:

df <- data.frame(x = 1:5)
p <- ggplot(df, aes(x, 1)) + geom_tile(aes(fill = x), colour = "white")
p

p + scale_fill_gradient(limits = c(2, 4))

p + scale_fill_gradient(limits = c(2, 4), oob = scales::squish)

6.6 Scales toolbox

6.6.1 Continuous position scales

# Convert from fuel economy to fuel consumption
ggplot(mpg, aes(displ, hwy)) +
  geom_point() +
  scale_y_continuous(trans = "reciprocal")

# Log transform x and y axes
ggplot(diamonds, aes(price, carat)) +
  geom_bin2d() +
  scale_x_continuous(trans = "log10") +
  scale_y_continuous(trans = "log10")

base <- ggplot(economics, aes(date, psavert)) +
  geom_line(na.rm = TRUE) +
  labs(x = NULL, y = NULL)
base # Default breaks and labels

## Warning in as.POSIXlt.POSIXct(x): unknown timezone 'default/America/
## New_York'

base + scale_x_date(date_labels = "%y", date_breaks = "5 years")

base + scale_x_date(
  limits = as.Date(c("2004-01-01", "2005-01-01")),
  date_labels = "%b %y",
  date_minor_breaks = "1 month"
  )

base + scale_x_date(
  limits = as.Date(c("2004-01-01", "2004-06-01")),
  date_labels = "%m/%d",
  date_minor_breaks = "2 weeks"
  )

6.6.2 Color

We’ll use a modern attempt called the hcl colour space, which has three components of h ue, c hroma and l uminance:

Hue is a number between 0 and 360 (an angle) which gives the “colour” of the colour: like blue, red, orange, etc.

Luminance is the lightness of the colour. A luminance of 0 produces black, and a luminance of 1 produces white.

Chroma is the purity of a colour. A chroma of 0 is grey, and the maximum value of chroma varies with luminance.

6.6.2.1 Continuous

erupt <- ggplot(faithfuld, aes(waiting, eruptions, fill = density)) +
  geom_raster() +
  scale_x_continuous(NULL, expand = c(0, 0)) +
  scale_y_continuous(NULL, expand = c(0, 0)) +
  theme(legend.position = "none")

erupt

erupt + scale_fill_gradient(low = "white", high = "black")

erupt + scale_fill_gradient(
  low = munsell::mnsl("5G 9/2"),
  high = munsell::mnsl("5G 6/8")
  )

mid <- median(faithfuld$density)
erupt + scale_fill_gradient2(midpoint = mid)

erupt + scale_fill_gradientn(colours = terrain.colors(7))

erupt + scale_fill_gradientn(colours = colorspace::heat_hcl(7))

erupt + scale_fill_gradientn(colours = colorspace::diverge_hcl(7))

erupt + scale_fill_distiller()

erupt + scale_fill_distiller(palette = "RdPu")

erupt + scale_fill_distiller(palette = "YlOrBr")

df <- data.frame(x = 1, y = 1:5, z = c(1, 3, 2, NA, 5))
p <- ggplot(df, aes(x, y)) + geom_tile(aes(fill = z), size = 5)
p

# Make missing colours invisible
p + scale_fill_gradient(na.value = NA)

# Customise on a black and white scale
p + scale_fill_gradient(low = "black", high = "white", na.value = "red")

6.6.2.2 Discrete

df <- data.frame(x = c("a", "b", "c", "d"), y = c(3, 4, 1, 2))
bars <- ggplot(df, aes(x, y, fill = x)) + 
  geom_bar(stat = "identity") +
  labs(x = NULL, y = NULL) + 
  theme(legend.position = "none")

bars

bars + scale_fill_hue(c = 40)

bars + scale_fill_hue(h = c(180, 300))

bars + scale_fill_brewer(palette = "Set1")

bars + scale_fill_brewer(palette = "Set2")

bars + scale_fill_brewer(palette = "Accent")

bars + scale_fill_grey()

bars + scale_fill_grey(start = 0.5, end = 1)

bars + scale_fill_grey(start = 0, end = 0.5)

#Bright colours work best with points
df <- data.frame(x = 1:3 + runif(30), y = runif(30), z = c("a", "b", "c"))
point <- ggplot(df, aes(x, y)) +
  geom_point(aes(colour = z)) +
  theme(legend.position = "none") +
  labs(x = NULL, y = NULL)
point + scale_colour_brewer(palette = "Set1")

point + scale_colour_brewer(palette = "Set2")

point + scale_colour_brewer(palette = "Pastel1")

# Subtler colours work better with areas
df <- data.frame(x = 1:3, y = 3:1, z = c("a", "b", "c"))
area <- ggplot(df, aes(x, y)) +
  geom_bar(aes(fill = z), stat = "identity") +
  theme(legend.position = "none") +
  labs(x = NULL, y = NULL)
area + scale_fill_brewer(palette = "Set1")

area + scale_fill_brewer(palette = "Set2")

area + scale_fill_brewer(palette = "Pastel1")

6.6.3 The manual discrete scale

plot <- ggplot(msleep, aes(brainwt, bodywt)) +
  scale_x_log10() +
  scale_y_log10()
plot

plot +
  geom_point(aes(colour = vore)) +
  scale_colour_manual(
    values = c("red", "orange", "green", "blue"),
    na.value = "grey50"
    )

## Warning: Removed 27 rows containing missing values (geom_point).

colours <- c(
  carni = "red",
  insecti = "orange",
  herbi = "green",
  omni = "blue"
  )
plot +
  geom_point(aes(colour = vore)) +
  scale_colour_manual(values = colours)

## Warning: Removed 32 rows containing missing values (geom_point).

huron <- data.frame(year = 1875:1972, level = as.numeric(LakeHuron))
ggplot(huron, aes(year)) +
  geom_line(aes(y = level + 5), colour = "red") +
  geom_line(aes(y = level - 5), colour = "blue")

ggplot(huron, aes(year)) +
  geom_line(aes(y = level + 5, colour = "above")) +
  geom_line(aes(y = level - 5, colour = "below"))

ggplot(huron, aes(year)) +
  geom_line(aes(y = level + 5, colour = "above")) +
  geom_line(aes(y = level - 5, colour = "below")) +
  scale_colour_manual("Direction",
                      values = c("above" = "red", "below" =
                                   "blue")
                      )

6.6.4 The identity scale

ggplot(luv_colours, aes(u, v)) +
  geom_point(aes(colour = col), size = 3) +
  scale_color_identity() +
  coord_equal()

Chapter 7 Positioning

Introduction

7.2 Faceting

facet_null() a single plot

7.2.1 facet_wrap()

as.table controls whether the facets are laid out like a table (TRUE), with highest values at the bottom-right, or a plot (FALSE), with the highest values at the top-right. dir controls the direction of wrap: horizontal or vertical.

library(ggplot2)
mpg2 <- subset(mpg, cyl != 5 & drv %in% c("4", "f") & class != "2seater")
base <- ggplot(mpg2, aes(displ, hwy)) +
  geom_blank() +
  xlab(NULL) +
  ylab(NULL)
base + facet_wrap(~class, ncol = 3)

base + facet_wrap(~class, ncol = 3, as.table = FALSE)

base + facet_wrap(~class, nrow = 3)

base + facet_wrap(~class, nrow = 3, dir = "v")

7.2.2 facet_grid()

base + facet_grid(. ~ cyl)

base + facet_grid(drv ~ .)

base + facet_grid(drv ~ cyl)

7.2.3 Controlling scales

p <- ggplot(mpg2, aes(cty, hwy)) +
  geom_abline() +
  geom_jitter(width = 0.1, height = 0.1)
p + facet_wrap(~cyl)

p + facet_wrap(~cyl, scales = "free")

economics_long

## # A tibble: 2,870 x 4
##          date variable value      value01
##        <date>   <fctr> <dbl>        <dbl>
##  1 1967-07-01      pce 507.4 0.0000000000
##  2 1967-08-01      pce 510.5 0.0002660008
##  3 1967-09-01      pce 516.3 0.0007636797
##  4 1967-10-01      pce 512.9 0.0004719369
##  5 1967-11-01      pce 518.1 0.0009181318
##  6 1967-12-01      pce 525.8 0.0015788435
##  7 1968-01-01      pce 531.5 0.0020679418
##  8 1968-02-01      pce 534.2 0.0022996199
##  9 1968-03-01      pce 544.9 0.0032177517
## 10 1968-04-01      pce 544.6 0.0031920097
## # ... with 2,860 more rows

ggplot(economics_long, aes(date, value)) + 
  geom_line() +
  facet_wrap(~variable, scales = "free_y", ncol = 1)

mpg2$model <- reorder(mpg2$model, mpg2$cty)
mpg2$manufacturer <- reorder(mpg2$manufacturer,-mpg2$cty)
ggplot(mpg2, aes(cty, model)) +
  geom_point() +
  facet_grid(manufacturer ~ ., scales = "free", space = "free") +
  theme(strip.text.y = element_text(angle = 0))

7.2.4 Missing faceting variables

df1 <- data.frame(x = 1:3, y = 1:3, gender = c("f", "f", "m"))
df2 <- data.frame(x = 2, y = 2)
ggplot(df1, aes(x, y)) +
  geom_point(data = df2, colour = "red", size = 2) +
  geom_point() +
  facet_wrap(~gender)

7.2.5 Gruoping vs. faceting

library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

df <- data.frame(
  x = rnorm(120, c(0, 2, 4)),
  y = rnorm(120, c(1, 2, 1)),
  z = letters[1:3]
)
ggplot(df, aes(x, y)) +
  geom_point(aes(colour = z))

ggplot(df, aes(x, y)) +
  geom_point() +
  facet_wrap(~z)

df_sum <- df %>%
  group_by(z) %>%
  summarise(x = mean(x), y = mean(y)) %>%
  rename(z2 = z)

ggplot(df, aes(x, y)) +
  geom_point() +
  geom_point(aes(colour = z2), data = df_sum, size = 4) +
  facet_wrap(~z)

# put all the data in the background of each panel:
df2 <- select(df, -z)
ggplot(df, aes(x, y)) +
  geom_point(data = df2, colour = "grey70") +
  geom_point(aes(colour = z)) +
  facet_wrap(~z)

7.2.6 Continuous variables

# Bins of width 1
mpg2$disp_w <- cut_width(mpg2$displ, 1)
# Six bins of equal length
mpg2$disp_i <- cut_interval(mpg2$displ, 6)
# Six bins containing equal numbers of points
mpg2$disp_n <- cut_number(mpg2$displ, 6)

plot <- ggplot(mpg2, aes(cty, hwy)) +
  geom_point() +
  labs(x = NULL, y = NULL)
plot + facet_wrap(~disp_w, nrow = 1)

plot + facet_wrap(~disp_i, nrow = 1)

plot + facet_wrap(~disp_n, nrow = 1)

7.3 Coordinate systems

There are two types of coordinate system. Linear coordinate system preserve the shape of geoms:

coord_cartesian(): the default Cartesian coordinate system, where the 2d position of an element is given by the combination of the x and y positions.

coord_fixed(): Cartesian coordinate system with a fixed aspect ratio.

coord_flip(): Cartesian coordinate system with x and y axes flipped.

On the other hand, non-linear coordinate can change the shapes: a straight line may no longer be straight. The closest distance between two points is no longer be a straight line.

coord_map()/coord_quickmap(): Map projections.

coord_polar(): Polar coordinates.

coord_trans(): Apply arbitrary transformations to x and y positions, after the data has been processed by the stat.

7.4 Linear coordinate systems

7.4.1 Zooming into a plot with coord_cartesian()

The key difference is how the limits work: when setting scale limits, any data outside the limits is thrown away; but when setting coordinate system limits we still use all the data, but we only display a small region of the plot. Setting coordinate system limits is like looking at the plot under a magnifying glass.

base <- ggplot(mpg, aes(displ, hwy)) +
  geom_point() +
  geom_smooth()
# Full dataset
base

## `geom_smooth()` using method = 'loess'

# Scaling to 5--7 throws away data outside that range
base + scale_x_continuous(limits = c(5, 7))

## `geom_smooth()` using method = 'loess'

## Warning: Removed 196 rows containing non-finite values (stat_smooth).

## Warning: Removed 196 rows containing missing values (geom_point).

# Zooming to 5--7 keeps all the data but only shows some of it
base + coord_cartesian(xlim = c(5, 7))

## `geom_smooth()` using method = 'loess'

7.4.2 Flipping the axes with coord_flip()

ggplot(mpg, aes(displ, cty)) +
  geom_point() +
  geom_smooth()

## `geom_smooth()` using method = 'loess'

# Exchanging cty and displ rotates the plot 90 degrees, but the smooth
# is fit to the rotated data.
ggplot(mpg, aes(cty, displ)) +
  geom_point() +
  geom_smooth()

## `geom_smooth()` using method = 'loess'

# coord_flip() fits the smooth to the original data, and then rotates
# the output
ggplot(mpg, aes(displ, cty)) +
  geom_point() +
  geom_smooth() +
  coord_flip()

## `geom_smooth()` using method = 'loess'

7.4.3 Equal scales with coord_equal()

coord_equal() fixes the ratio of length on the x and y axes. The default ratio, ensures that the x and y axes have equal scales

7.5 Non-linear coordinate systems

rect <- data.frame(x = 50, y = 50)
line <- data.frame(x = c(1, 200), y = c(100, 1))
base <- ggplot(mapping = aes(x, y)) +
  geom_tile(data = rect, aes(width = 50, height = 50)) +
  geom_line(data = line) +
  xlab(NULL) + ylab(NULL)

## Warning: Ignoring unknown aesthetics: width, height

base

base + coord_polar("x")

base + coord_polar("y")

base + coord_flip()

base + coord_trans(y = "log10")

base + coord_equal()

Firstly, the parameterisation of each geom is changed to be purely location-based, rather than location and dimension based.

df <- data.frame(r = c(0, 1), theta = c(0, 3 / 2 * pi))
ggplot(df, aes(r, theta)) +
  geom_line() +
  geom_point(size = 2, colour = "red")

interp <- function(rng, n) {
  seq(rng[1], rng[2], length = n)
  }

munched <- data.frame(
  r = interp(df$r, 15),
  theta = interp(df$theta, 15)
  )
ggplot(munched, aes(r, theta)) +
  geom_line() +
  geom_point(size = 2, colour = "red")

transformed <- transform(munched,
                         x = r * sin(theta),
                         y = r * cos(theta)
                         )

 ggplot(transformed, aes(x, y)) +
     geom_path() +
     geom_point(size = 2, colour = "red") +
     coord_equal()

7.5.1 Transformations with coord_trans()

# Linear model on original scale is poor fit
base <- ggplot(diamonds, aes(carat, price)) +
  stat_bin2d() +
  geom_smooth(method = "lm") +
  xlab(NULL) +
  ylab(NULL) +
  theme(legend.position = "none")
base

# Better fit on log scale, but harder to interpret
base +
  scale_x_log10() +
  scale_y_log10()

# Fit on log scale, then backtransform to original.
# Highlights lack of expensive diamonds with large carats
pow10 <- scales::exp_trans(10)
base +
  scale_x_log10() +
  scale_y_log10() +
  coord_trans(x = pow10, y = pow10)

7.5.2 Polor coordinates with coord_polar()

base <- ggplot(mtcars, aes(factor(1), fill = factor(cyl))) +
  geom_bar(width = 1) +
  theme(legend.position = "none") +
  scale_x_discrete(NULL, expand = c(0, 0)) +
  scale_y_continuous(NULL, expand = c(0, 0))
# Stacked barchart
base

# Pie chart
base + coord_polar(theta = "y")

# The bullseye chart
base + coord_polar()

7.5.3 Map projections with coord_map()

#library(map_data)
# Prepare a map of NZ
nzmap <- ggplot(map_data("nz"), aes(long, lat, group = group)) +
  geom_polygon(fill = "white", colour = "black") +
  xlab(NULL) + 
  ylab(NULL)
# Plot it in cartesian coordinates
nzmap

# With the aspect ratio approximation
nzmap + coord_quickmap()

world <- map_data("world")
worldmap <- ggplot(world, aes(long, lat, group = group)) +
  geom_path() +
  scale_y_continuous(NULL, breaks = (-2:3) * 30, labels = NULL) +
  scale_x_continuous(NULL, breaks = (-4:4) * 45, labels = NULL)


worldmap + coord_map()

   # Some crazier projections
worldmap + coord_map("ortho")

worldmap + coord_map("stereographic")